xu::dfloat stores a decimal number with 18 digits of precision and exponent between [-128, 127)
Basic arithmetic (+ - * /) support. All operations are truncating (round towards zero) to 18 digits.
Requires GCC's __uint128_t. Uses fixed-width integer types internally.
Also requires C++11.
License information can be found in the LICENSE file.
A rough test suggests that xu::dfloat is slower than double by a factor on the order of 20:1 on a x86_64 machine.